Abstract
On a widespread reading, the Carnap-Quine debate about ontology concerns the objectivity and non-triviality of ontological claims. I argue that this view mischaracterizes Carnap’s aims in “Empiricism, Semantics, and Ontology” (ESO): Carnap’s fundamental goal is to free up decisions about scientific language from constraints deriving from ontological doctrine. The contention, based on his internal/external (i/e) distinction, that ontological claims are either meaningless or trivial was Carnap’s means to achieving this more fundamental goal. Setting the record straight on this point brings out three important and often overlooked features of Carnap’s views on ontology. First, the target of Carnap’s critique in ESO is not Quine’s mature views on ontology, as laid out in “On What There Is”. Rather, Carnap is responding to arguments for nominalism that were given by Tarski, Goodman, and Quine in the 1940s. Second, a more general rejection of conservatism in theory choice is essential to Carnap’s aims and is implicit in his fundamental views on language. Third, even if it turns out to be tenable, the i/e distinction is not adequate for Carnap’s aim in ESO. Drawing on his basic conception of scientific language, I will suggest an alternative approach on his behalf.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
I use ‘language for science’ and ‘linguistic framework’ interchangeably. Carnap may have intended for a linguistic framework to be a particular kind of fragment of a language for science. However, since he drops the term in favor of ‘language’ in his (1963b, 868–73), I think it reasonable to assume that any difference between the two terms is unimportant for his views.
Another aim is to show that reference to abstract objects is compatible with empiricism. I briefly return to this concern in §5.
While Carnap denies that decisions about the language for science are subject to theoretical justification, he believes that they are subject to practical justification, that is, that they can be more or less useful for their chosen purpose. As I discuss in §7 Carnap’s chosen purpose for languages for science is the formulation and derivation of observational predictions and reports.
Carnap (1963b, 873) illustrates the target of his critique of ontology through the example of an argument of this kind against a language that posits classes of classes.
Barry Stroud (1984) also understands Carnapian linguistic frameworks in terms of an empiricist criterion of cognitive significance. However, on Stroud’s understanding of this criterion, empirical confirmability is necessary for cognitive significance; see his (1984, 171). The account of Carnap’s empiricist criterion I gave above, on which analytic truth or falsity is sufficient for cognitive significance, is standard, and is supported by a preponderance of textual evidence; see my exegetical defense of the interpretation later in this section.
Frameworks whose questions are answerable by empirical methods include the “world of things” (Carnap 1956a, 206–8) and the “system of thing properties” (Carnap 1956a, 211–2). Frameworks whose questions have analytic answers include the “system of numbers” (Carnap 1956a, 208–9), the “system of propositions” (Carnap 1956a, 209–11), and the “spatio-temporal coordinate system for physics” (Carnap 1956a, 212–3).
Carnap sometimes takes various frameworks to contain their own distinctive kind of typed variable. Seizing on these passages, Quine argues that the i/e distinction “is of little concern to us apart from the adoption of something like the theory of types” (1976b, 132). However, Carnap later draws the distinction with respect to a language for classes that contains “one kind of variable, which are not type-restricted” (1963b, 872). For further objections to Quine’s interpretation of the i/e distinction, see Bird (2003).
It is worth noting another misinterpretation running through Quine’s (1976b) criticisms of ESO. Quine read Carnap as maintaining that we can use language that is ontologically committed, in Quine’s sense, to abstract objects (among other things) without “really” positing such objects. While there are passages in ESO that encourage this reading, I am convinced by Alspector-Kelly’s (2001) objections to it.
I assume that Carnap’s idea of internal questions being trivial is what motivates Eklund’s (2009) attribution to him of the thesis that they are shallow; Eklund does not cite textual evidence or any specific aspect of Carnap’s views on behalf of this attribution.
Is this real freedom? Would anyone ever want to exercise it? Carnap (1959, 79-80) thinks so. He argues that lyrical poetry and traditional metaphysics are both often composed of cognitively meaningless inscriptions. But this doesn’t mean the lyrical poet is doing anything wrong. Carnap sees poetry as an “expression” of the author’s attitudes towards life—to be contrasted with an assertion about her attitudes—and believes that non-cognitive discourse can be an effective vehicle for this kind of expression. The metaphysician, by contrast, gives arguments and attempts to assert theses. It is this combination of cognitive goals with non-cognitive language in traditional metaphysics that Carnap opposes. Traditional metaphysics, according to Carnap, “pretends to be something that it is not” (1959, 79).
Many years later, in Word and Object, Quine maintains that, if we are being conscientious about our methodology, empirical adequacy and simplicity should trump conservatism (1960, 20–1). On this approach, conservatism is called on only to decide between equally viable theories, between which we are otherwise indifferent. Carnap would not oppose settling ties this way any more than by coin toss.
The only examples of the wrong approach to ontology that Carnap gives are Bernays (1935) and Nagel (1948). He does not discuss either paper in any detail. However, it is clear that Carnap sees the approach to ontology that he wants to rebut as widespread. For this reason, his failure to cite Goodman and Quine (1947) is not, in my view, strong evidence that this paper was not among his intended targets in ESO.
Carnap uses the example of numbers to illustrate how sentences like this one may be derived from the semantical rules for the framework. He believes that statements of the form ⌜⌜x⌝ designates x⌝ are analytic, “provided the term [⌜x⌝] is a constant in an accepted framework” (1956a, 217). Therefore, “red’ designates red’ is analytic in the framework of thing properties. Carnap claims, further, that since the framework of properties contains the general term ‘property’, ‘red is a property’ is also analytic. The two analytic sentences just mentioned jointly imply “red’ designates a property’, and so this sentence is likewise analytic within the framework.
For example, ‘There are more cats than dogs’ becomes ‘Every individual that contains a bit of each cat is bigger than some individual that contains a bit of each dog’ (Goodman and Quine 1947, 109–110).
An anonymous referee for this journal suggested that Goodman and Quine’s remarks amount to motivation, as opposed to an argument, for their nominalistic language. Whatever this difference comes to, as the referee rightly points out, Carnap’s response would be the same.
The “philosophical intuition” appealed to here is plausibly what Carnap has in mind when he dismisses “ontological insight” (1956a, 221) as a basis for language choice.
Goodman and Quine give a secondary and independent argument, stating that their renunciation of abstract objects is “fortified” by the fact that “[e]scape from [the class-theoretic] paradoxes can apparently be effected only by recourse to alternative rules whose artificiality and arbitrariness arouse suspicion that we are lost in a world of make-believe” (1947, 105).
There are two other objections to empiricist criteria of cognitive significance that have been particularly influential. The one holds that advocacy of such criteria is itself neither empirically significant nor analytically true or false, and is therefore, by its own lights, cognitively meaningless. The other holds that empiricist criteria of cognitive significance presuppose an untenable distinction between observation and theoretical language. These objections lie outside the scope of this essay.
An objection to this line of interpretation is that without an a/s distinction, Quine’s conceptual scheme cannot count as a linguistic framework and so the statements belonging to it cannot be internal.
There are three possible responses to this objection. The first, and most direct, maintains that the statements of a linguistic framework need not be divided into analytic and synthetic classes. A linguistic framework is composed of sentences that are either empirically significant or analytic; so a language that is composed entirely of empirically significant sentences is composed of internal sentences, even if we eschew meta-linguistic analyticity and syntheticity predicates.
The second response is to argue that Carnap should, in light of the controversies it has generated, abandon his empiricist criterion of cognitive significance, including the a/s distinction. In its place, he could take a pragmatic approach to delimiting the class of empiricist languages and drawing the i/e distinction. I briefly discuss this option in §§6–7. The content of Carnap’s pragmatism (§8) would ensure that “the simplest conceptual scheme into which the disordered fragments of raw experience can be fitted and arranged” (Quine 1980a, 16) would count as an empiricist language containing internal statements.
A further objection to the claim that Quine’s rejection of abstract objects could be internal holds that, to be internal, the statement ‘There are no abstract objects’ would have to be analytic within the relevant linguistic frameworks. So without an a/s distinction, Quine could not have been employing such a framework. In response, I note, first, that I know of no textual evidence that Carnap would have thought that the statement in question must be analytic. Second, there is reason to think it could be empirically significant. Two historically influential (if controversial) criteria for abstract objecthood are absence of spatial location and lack of causal efficacy (Rosen 2014). But the concepts in terms of which both of these criteria are stated would plausibly be, for Carnap, combinations of logical and empirically significant concepts (negation and spatial location/causal efficacy). And, by Carnap’s criterion, a sentence that is composed of logical and empirically significant terms is itself empirically significant (Carnap 1956c, 60).
Thanks to an anonymous referee for this journal for clarifying this aspect of Quine’s views.
This may seem like a counterintuitive conception of science. One might think that we rather use scientific language to correctly describe the world. The more basic motivation that leads Carnap away from this view and towards his pragmatism is, very briefly, a desire to restrict scientific discourse to fruitful questions. A full discussion of this more basic motivation lies outside the scope of this essay.
See also Ricketts (1982, 121).
This quotation comes from a passage in which Carnap asserts that “[a]ppeal to ontological insight will not carry much weight” (1956a, 221). As an anonymous referee for this journal has pointed out, this assertion might be taken to suggest that, contrary to my interpretation, ontological insight does carry some weight. I believe that, as this referee suggests, Carnap’s use of ‘much’ here is simply unnecessary. Earlier in the paper, Carnap contrasts the view on which adding a new entity to our ontology “is legitimate only if it can be justified by an ontological insight supplying an affirmative answer to the question of reality” with his view that “the introduction of the new ways of speaking does not need any theoretical justification because it does not imply any assertion of reality” (1956a, 214). The upshot of these remarks is that, not only is ontological insight unnecessary for adding an entity to our ontology, but such insights do not justify such an addition at all—the decision to add to our ontology is not the kind of thing that is subject to the kind of justification that ontological insights putatively provide.
Quine (1960, 20) overlooks this distinction.
While Sider somewhat downplays the role of match with belief in this passage, it is worth noting that he leaves open the possibility that it can outweigh other desiderata. Sider (2001, xv-xvi) does not downplay the role of match with belief in the same way, writing “[o]ne approaches metaphysical inquiry with a number of beliefs…. One then develops a theory preserving as many of these ordinary beliefs as possible, while remaining consistent with science.”
References
Alspector-Kelly, M. (2001). On quine on Carnap on ontology. Philosophical Studies, 102(1), 93–122.
Bernays, P. (1935). Sur le Platonisme Dans les Mathématiques. L’Enseignement Mathématique, 34, 52–69.
Bird, G. (2003). Carnap’s internal and external questions. In T. Bonk (Ed.), Language, truth, and knowledge: contributions to the philosophy of Rudolf Carnap (pp. 97–131). Norwell: Kluwer.
Carnap, R. (1936). Testability and Meaning. Philosophy of Science, 3–4, 410–71 ,1–40.
Carnap, R. (1937). The Logical Syntax of Language. (a. Smeaton, trans.). New York: Open Court.
Carnap, R. (1939). Foundations of Logic and Mathematics (Vol. I of The International Encyclopedia of Unified Science). Chicago: University of Chicago Press.
Carnap, R. (1953). Formal and factual science. In H. Feigl & M. Brodbeck (Eds.), Readings in the Philosophy of Science (H. Feigl, & M. Brodbeck, trans) (pp. 123–128). New York: Appleton-Century-Crofts.
Carnap, R. (1956a). Empiricism, semantics, and ontology. In R. Carnap (Ed.), Meaning and necessity: a study in semantics and modal logic (pp. 205–221). Chicago: University of Chicago Press.
Carnap, R. (1956b). Meaning and Synonymy in Natural Languages. In Meaning and Necessity: A Study in Semantics and Modal Logic (p. 233). Chicago: The University of Chicago Press.
Carnap, R. (1956c). The Methodological Character of Theoretical Concepts. (H. Feigl, & M. Scriven, Eds.) Minnesota Studies in the Philosophy of Science, The Foundations of Science and the Concepts of Psychology and Psychoanalysis, I, 38–76.
Carnap, R. (1956d). Meaning and necessity: a study in semantics and modal logic. Chicago: University of Chicago Press.
Carnap, R. (1959). The Elimination of Metaphysics Through the Logical Analysis of Language. A. J. Ayer, Ed., & A. Pap, Trans. (pp. 60–81). New York: The Free Pree, Co.
Carnap, R. (1963a). Intellectual autobiography. In P. A. Schilpp (Ed.), The philosophy of Rudolf Carnap (pp. 3–84). LaSalle: Open Court.
Carnap, R. (1963b). Replies and systematic expositions. In P. A. Schilpp (Ed.), The philosophy of Rudolf Carnap (pp. 859–999). LaSalle: Open Court.
Carnap, R. (1987). On Protocol Sentences. Noûs, 21(4), 457–470.
Carnap, R. (1991). Quine on analyticity. In R. Creath (Ed.), Dear Carnap, dear van: the quine-Carnap correspondence and related work (pp. 427–432). Berkeley: The University of California Press.
Carnap, R. (2003). The Logial Structure of the World and Pseudoproblems in Philosophy. (R. George, trans.). Chicago: Open Court.
Carus, A. (2003). Sellars, Carnap, and the logical space of reasons. In S. Awodey & C. Klein (Eds.), Carnap brought home: the view from Jena (pp. 317–355). New York: Open Court.
Carus, A. (2007). Carnap and twentieth century thought: explication as enlightenment. New York: Cambridge University Press.
Chalmers, D. (2009). Ontological Anti-Realism. In D. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: new essays on the foundations of ontology (pp. 77–129). New York: Oxford University Press.
Creath, R. (Ed.) (1986). Dear Carnap, dear van. The quine-Cornap correspondence and related work. Berkeley: University of California Press.
Creath, R. (1992). Carnap’s conventionalism. Synthèse, 93(1–2), 141–166.
Eklund, M. (2009). Carnap and ontological pluralism. In D. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: new essays on the foundations of ontology (pp. 130–156). New York: Oxford University Press.
Eklund, M. (2013). Carnap’s Metaontology. Noûs, 47(2), 229–249.
Fine, K. (2009). The question of ontology. In D. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: new essays on the foundations of ontology (pp. 157–177). New York: Oxford University Press.
Frost-Arnold, G. (2013). Carnap, Tarski, and quine at Harvard: conversations on logic, mathematics, and science. Chicago: Open Court.
Goldfarb, W., & Ricketts, T. (1992). Carnap and the Philosophy of Mathematics. In Science and Subjectivity: The Vienna Circle and Twentieth Century Philosophy (pp. 61–78). Berlin: Akadamie-Verlag.
Haack, S. (1976). Some preliminaries to ontology. Journal of Philosophical Logic, 5(4), 457–454.
Mancosu, P. (2005). Harvard 1940-1941: Tarski, Carnap and quine on a finitistic language of mathematics for science. History and Philosophy of Logic, 26, 327–357.
Manley, D. (2009). Introduction: a guided tour of Metametaphysics. In D. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: new essays on the foundations of ontology (pp. 1–37). New York: Oxford University Press.
Nagel, E. (1948). Review of Meaning and Necessity. The Journal of Philosophy, 45(17), 467–472.
Quine, W. (1943). Notes on existence and necessity. Journal of Philosophy, 40, 113–127.
Quine, W. (1947). On universals. The Journal of Symbolic Logic, 12(3), 74–84.
Quine, W. (1960). Word and object. Cambridge: MIT Press.
Quine, W. (1969). Speaking of Objects. In Ontological Relativity and Other Essays (pp. 1–25). New York: Columbia University Press.
Quine, W. (1976a). Carnap and Logical Truth. In The Ways of Paradox and Other Essays (p. 107). Cambridge: Harvard University Press.
Quine, W. (1976b). On Carnap’s views on ontology. In W. Quine (Ed.), The ways of paradox (Revised ed., pp. 126–134). Cambridge: Harvard University Press.
Quine, W. (1980a). On What There Is. In From a Logical Point of View: Nine Logico-Philosophical Essays (Second, Revised ed., pp. 1–19). Cambridge: Harvard University Press.
Quine, W. (1980b). Two Dogmas of Empiricism. In From a Logical Point of View: Nine Logico-Philosophical Essays (pp. 26–52). Cambridge: Harvard University Press.Second, Revised ed
Quine, W., & Goodman, N. (1947). Steps toward a constructive nominalism. The Journal of Symbolic Logic, 12(4), 105–122.
Ricketts, T. (1982). Rationality, translation, and epistemology naturalized. The Journal of Philosophy, 79(3), 117–136.
Ricketts, T. (1994). Carnap’s principle of tolerance, empiricism, and conventionalism. In B. Hale & P. Clark (Eds.), Reading Putnam (pp. 176–200). Oxford: Blackwell.
Rosen, G. (2014, Fall). Abstract Objects. (E. N. Zalta, Editor) Retrieved 5 Oct 2014, from The Stanford Encyclopedia of Philosophy: <http://plato.stanford.edu/archives/fall2014/entries/abstract-objects/>
Sider, T. (2001). Four Dimensionalism: an ontology of persistence and time. New York: Oxford University Press.
Sider, T. (2009). Ontological Realism. In D. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: new essays on the foundations of ontology (pp. 384–423). New York: Oxford University Press.
Stein, H. (1992). Was Carnap entirely wrong, after all? Synthèse, 93, 275–295.
Stroud, B. (1984). The significance of philosophical Scepticism. New York: Oxford University Press.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interest
Research done for this essay was not supported by any funding and did not involve any potential conflicts of interest.
Human and Animal Rights
Research for this essay did not involve any human or animal participants.
Informed Consent
N/A
Rights and permissions
About this article
Cite this article
Surovell, J. Language, Ontology, and the Carnap-Quine Debate. Philosophia 45, 811–833 (2017). https://doi.org/10.1007/s11406-016-9795-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11406-016-9795-7